Conventionally, an embroidery data generator is known which is capable of automatically determining the stitch patterns to be applied to multiple sub-patterns that constitute an embroidery pattern. Examples of stitch pattern include a satin stitch, a fill stitch, and a running stitch. In one example of an embroidery data generator, a circularity index is calculated using a predetermined calculation formula based on the area and the perimeter of a sub-pattern. The circularity index is a numerical value indicating whether the sub-pattern is circular or elongate. The embroidery data generator automatically determines the stitch pattern to be applied to the sub-pattern by comparing the calculated circularity index with the preset threshold value.